trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.9998530492285083,
  "global_step": 13608,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 3.6737692872887585e-08,
      "loss": 3.3894,
      "step": 1
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.8368846436443794e-05,
      "loss": 3.3892,
      "step": 500
    },
    {
      "epoch": 0.22,
      "learning_rate": 3.673769287288759e-05,
      "loss": 3.4112,
      "step": 1000
    },
    {
      "epoch": 0.33,
      "learning_rate": 4.9432514085082064e-05,
      "loss": 3.4369,
      "step": 1500
    },
    {
      "epoch": 0.44,
      "learning_rate": 4.739119784437005e-05,
      "loss": 3.4317,
      "step": 2000
    },
    {
      "epoch": 0.55,
      "learning_rate": 4.534988160365804e-05,
      "loss": 3.4256,
      "step": 2500
    },
    {
      "epoch": 0.66,
      "learning_rate": 4.330856536294603e-05,
      "loss": 3.4214,
      "step": 3000
    },
    {
      "epoch": 0.77,
      "learning_rate": 4.127133175471544e-05,
      "loss": 3.4182,
      "step": 3500
    },
    {
      "epoch": 0.88,
      "learning_rate": 3.9230015514003435e-05,
      "loss": 3.4124,
      "step": 4000
    },
    {
      "epoch": 0.99,
      "learning_rate": 3.718869927329142e-05,
      "loss": 3.4087,
      "step": 4500
    },
    {
      "epoch": 1.1,
      "learning_rate": 3.51473830325794e-05,
      "loss": 3.4057,
      "step": 5000
    },
    {
      "epoch": 1.21,
      "learning_rate": 3.31060667918674e-05,
      "loss": 3.399,
      "step": 5500
    },
    {
      "epoch": 1.32,
      "learning_rate": 3.1064750551155385e-05,
      "loss": 3.3956,
      "step": 6000
    },
    {
      "epoch": 1.43,
      "learning_rate": 2.9023434310443376e-05,
      "loss": 3.3923,
      "step": 6500
    },
    {
      "epoch": 1.54,
      "learning_rate": 2.6982118069731367e-05,
      "loss": 3.3893,
      "step": 7000
    },
    {
      "epoch": 1.65,
      "learning_rate": 2.4940801829019355e-05,
      "loss": 3.3858,
      "step": 7500
    },
    {
      "epoch": 1.76,
      "learning_rate": 2.2899485588307342e-05,
      "loss": 3.3829,
      "step": 8000
    },
    {
      "epoch": 1.87,
      "learning_rate": 2.085816934759533e-05,
      "loss": 3.3815,
      "step": 8500
    },
    {
      "epoch": 1.98,
      "learning_rate": 1.8820935739364743e-05,
      "loss": 3.3786,
      "step": 9000
    },
    {
      "epoch": 2.09,
      "learning_rate": 1.677961949865273e-05,
      "loss": 3.3752,
      "step": 9500
    },
    {
      "epoch": 2.2,
      "learning_rate": 1.4738303257940722e-05,
      "loss": 3.3681,
      "step": 10000
    },
    {
      "epoch": 2.31,
      "learning_rate": 1.269698701722871e-05,
      "loss": 3.3666,
      "step": 10500
    },
    {
      "epoch": 2.42,
      "learning_rate": 1.0655670776516699e-05,
      "loss": 3.365,
      "step": 11000
    },
    {
      "epoch": 2.54,
      "learning_rate": 8.618437168286112e-06,
      "loss": 3.3638,
      "step": 11500
    },
    {
      "epoch": 2.65,
      "learning_rate": 6.5771209275741e-06,
      "loss": 3.3629,
      "step": 12000
    },
    {
      "epoch": 2.76,
      "learning_rate": 4.535804686862089e-06,
      "loss": 3.362,
      "step": 12500
    },
    {
      "epoch": 2.87,
      "learning_rate": 2.4944884461500775e-06,
      "loss": 3.3622,
      "step": 13000
    },
    {
      "epoch": 2.98,
      "learning_rate": 4.531722054380665e-07,
      "loss": 3.3615,
      "step": 13500
    },
    {
      "epoch": 3.0,
      "step": 13608,
      "total_flos": 7.655079399630727e+19,
      "train_loss": 3.390678170251538,
      "train_runtime": 24048.3069,
      "train_samples_per_second": 1738.566,
      "train_steps_per_second": 0.566
    }
  ],
  "max_steps": 13608,
  "num_train_epochs": 3,
  "total_flos": 7.655079399630727e+19,
  "trial_name": null,
  "trial_params": null
}